Christine Milligan, PhD

Global clinical supply professionals are driven by a simple credo: to provide the highest possible quality clinical trial supplies in a fast and efficient manner. Given the scope of global clinical trial activity today, delivering on this credo is a tall order. There are approximately 40,000 unique investigators worldwide conducting at least one FDA-regulated clinical trial.

	There are a number of factors intensifying pressure on global clinical supply professionals. To mention a few key factors: the economics of supply chain management and distribution have grown substantially due to study complexity and increases in shipping costs, labor costs, technology solutions costs and changes in drug therapy properties requiring additional shipping and packaging considerations (e.g., cold chain and temperature sensitive shipping requirements; combination therapies; and companion diagnostics).

 A recent survey of 250 supply chain executives found that two-thirds anticipated significantly increasing spending on clinical trial logistics over the next two years.

	Investigative site workload has increased substantially, making it more difficult to receive operating support from study staff. A Tufts Center for the Study of Drug Development (Tufts CSDD) study examining protocol complexity and burden on clinical trial site staff, for example, found that investigative site work effort to administer each protocol had increased 64% between 2002 and 2012.

Short study start-up lead times, more complex study drug packaging requirements and tight study conduct durations place significant pressure on supply chain managers to provide efficient supply ordering and tracking solutions for investigative sites to use.

	Increasing demands from global regulations are also impacting the supply chain. In 2013, the European Union enacted good distribution practices (GDPs) which are becoming the adopted guidance globally.

 Although the timeline is still under discussion, new regulations for clinical trials conducted in the European Union will occur in October 2018.

 This new law could potentially impact the supply chain, although it is primarily geared to existing good manufacturing practices (GMPs) for products covered by an existing directive. 

	Clinical supply professionals face ongoing risks to the supply chain security, traceability and authentication of a product. There are constant threats of counterfeiting, which can impact patient safety.

 At the same time, clinical supply logistics managers face substantially higher levels of visibility and demand to play a more strategic role within their organizations. 

	Despite the heavy and increasing pressure on clinical supply chain professionals, there is little to no historical data characterizing performance, measuring the impact of new processes and solutions and identifying areas of improvement. In response, Tufts CSDD-in collaboration with a diverse group of pharmaceutical and

 biotech companies, clinical supply logistics providers and suppliers-conducted a global clinical supply logistics study. The study gathered survey data and clinical study performance data from companies in order to examine current management practices and strategies that have been implemented to enhance efficiency and productivity within the clinical supply logistics area. The results of the study suggest that companies are implementing a variety of strategies that, together, are assisting organizations in maintaining high quality and low error rates and identifying contributing factors for waste. All sponsor companies, however, are encountering wide variation in mean shipping times.

	Tufts CSDD convened a roundtable meeting with global clinical supply managers and directors to identify the most critical areas for which benchmark data could be gathered. Later, the group narrowed the list of topics to the focus of the current study. A total of 15 companies participated in the study, including Astellas, AstraZeneca, Biogen, Bristol-Myers Squibb, Eli Lilly, Janssen (a Johnson & Johnson company), Merck Serono, Pfizer, Sunovion, UCB, Catalent Pharma Solutions, Fisher Clinical Services, Clinigen Group, Medidata Solutions, and Endpoint Clinical. This working group collectively developed both a survey and a data collection instrument. The study aims were to gather quantitative metrics and to capture practices and strategies regarding clinical supply logistics. The study focused on warehousing and distribution and did not examine supply sourcing, manufacturing or packaging and labelling. 

	The study examined a number of areas within global supply logistics, including management of distribution strategies among organizations and measures of success of these strategies. Approaches to temperature monitoring and implementation of cost-reduction measures were also investigated. Lastly, use of interactive response technologies (IRT) and other tools used to increase inventory visibility and study supply forecasting were explored as part of the study.

	Tufts CSDD facilitated the working group process and collaborated on the development of the survey and data collection instrument. All data was gathered and analyzed by Tufts CSDD. The study was launched in the fall of 2014 and data was gathered through early 2015. A preliminary results meeting was held in London in June 2015 and a final results meeting was held in Boston in October 2015 with working group companies to discuss the results of the analyses.

	The survey gathered demographic data on respondent’s organization type and size, the top therapeutic areas in which clinical trials are being conducted at their organization, and countries to which their organization is sending drugs and supplies. Data was also gathered on depot locations and regional hubs; distribution networks and strategies; return management; temperature sensitive shipping, use of IRT; inventory and planning systems; and training and communication. The survey was sent via an email invitation to select company contacts as well as to the working group companies. In addition to the working group, there were 97 invitations sent out across the industry, to pharmaceutical companies, contract research organizations (CROs) and service providers, asking potential respondents to complete the survey. A total of 17 respondents completed the survey.

	In addition to the survey component, data on clinical supply logistics were gathered for recent clinical studies conducted by the working group companies. These data examined global studies conducted by companies in the past five years. Organizations were asked to contribute data from at least eight of their studies (two trials per phase) across a broad range of therapeutic areas.

	The data collection instrument gathered clinical study-specific cycle time and performance data from participating companies. Data was gathered on study characteristics, including study phase and status, therapeutic area, disease states, 

global sites and planned enrollment. Data on key metrics included number of shipments, on-time shipments, cycle times, product impact, waste, costs spent on study services and outsourced logistics services. After an interim results meeting, working group participants agreed to provide additional data on product impact, waste and shipping times based on refined definitions. Additional data was contributed by 12 working group companies. These data are included in the current analyses in this article.

	Seventeen companies responded to the logistics survey and 12 companies provided logistics data for 73 clinical studies across a diverse group of therapeutic areas, including oncology, neuroscience and central nervous system (CNS), immunology and respiratory studies. Of the 73 studies gathered, there were 14 (19%) in Phase 1, 19 (26%)  in Phase II, 31 (42%) in Phase III  and nine (12%) in Phase IV. (Some companies contributed less than two studies per phase as additional data were not available). The top therapeutic areas of studies gathered were oncology, CNS and neuroscience, immunology and respiratory (see Figure 1 below). For all study data gathered, companies sent drugs and supplies to 5,682 sites across all global regions. Shipments were distributed across North America, Western and Eastern Europe, Asia-Pacific, Latin American and “Rest of World.”

	The mean planned enrollment time of studies in Phase II and Phase III were 11.8 months and 16.3 months, respectively. Variance in enrollment time across studies from 11 therapeutic areas was small, ranging from 0.1 to 0.9 (see Figure 2 below). The variance in enrollment drives clinical supply logistics strategy and performance. The studies in CNS and neuroscience, oncology and metabolic and endocrine areas had high variances, while hematology and infectious disease had low variances.

	Survey respondents indicated that drugs and supplies were shipped across all regions with the largest proportion of shipments in North America (78%), followed by Western Europe (66%), Eastern Europe (63%), Latin America (29%), Asia-Pacific (22%) and Rest of World (18%). (Percentages represent percent of total shipments and do not add up to 100%). Companies reported that their top regional hubs are in Western Europe (38%), Asia-Pacific (31%) and the Rest of the World (13%). Top local depot locations are in Latin America (49%), North America (41%) and Asia-Pacific (36%) regions. Distribution strategies varied with 31 studies (42%) using a centralized approach, 30 (41%) were managed regionally and 12 (16%) were locally managed. A centralized approach is defined as one depot for a global study distributing to all sites. Regional hubs were defined as the main distribution hub responsible for shipping to each country within a wider region, and local depots were depots with regular shipping only to domestic locations.

	Respondents estimated that nearly 70% of studies were using IRT systems for multiple processes, including drug ordering, randomization and expiry data management. Data was gathered from 12 companies for this question. Other usage of IRT systems were in the areas of drug reconciliation and temperature excursion tracking (see Figure 3 below). The majority used IRT for drug ordering (12 of 12 companies), randomization (10 of 12) and data extension management (eight of 12). IRT is also being used for effective control of blinding and unblinding, as it hosts dosing assignment information, instead of paper-based control. The primary ways that companies envision the use of IRT evolving in their organization were through new functionalities and standardized modular platforms. Staff from clinical supply and clinical operations were identified as primary decision-makers in outsourcing or keeping IRT systems in-house. 

	The results of the survey also suggest that use of IRT and integration are two approaches that have increased inventory visibility within organizations. Being able to integrate multiple systems (for example, an integrated EDC/IRT approach) provides access to all sources of data. Survey respondents indicated that they used multiple systems, ranging from one to six, and the majority of these systems are outsourced. Some of the platforms were integrated with IRT and included EDC, drug accountability and enterprise resource planning (ERP).

	Many companies report IRT improvement initiatives in new systems or processes through standardization of requirements, interfacing with other clinical systems or revising system specification processes or functionalities. Also, cross-functional governance teams are in place at most organizations as part of improvement initiatives that oversee functionality standards and processes internally as well as with vendors. Lastly, IRT improvement initiatives for provider and partner evaluations were also being implemented. Examples of these initiatives included providing formal training and improving IRT standards among sponsors, CROs and IRT providers.

	For the 73 studies evaluated, there were 1,538 shipments, on average, made to investigative sites and 17 bulk shipments to in-country depots in our analyses. An overwhelming majority of shipments-97%-arrived on time to sites and 80% to in-country depots. We also found that in analyzing mean shipping time in days by distribution strategy that there was a wide variation and the longest times were found in centralized approaches (5.8 days in North America to 27.8 days in Rest of World) or one depot for a global study distributing to all sites (see Figure 4 below). Distribution strategies were divided among centralized (27 studies), local (11), and regional (19) approaches.

	The largest proportion of clinical supply logistics costs across all clinical studies analyzed were for courier and depot costs (49%) and storage and distribution costs (40%). Of the total studies, 43 were outsourced and 30 were run internally. A small percentage of total studies (3%) from the company data gathered experienced product impact with the greater number of reports involving errors in shipping and handling, including temperature excursion. Other errors included site mishandling, customs intervention and errors with IRT. Errors with packaging in labelling are among the least (see Figure 5 below).

	Product waste was examined across studies on three measures: the percent of total manufactured product packaged, percent of the total packaged product that was shipped to sites, and the percent of product shipped that was dispensed. The results indicated that 67.8% of the product shipped to sites was dispensed to patients based on 12 companies and 57 studies (see Figure 6 below). 

	Companies varied in their approaches to calculating supply overage. Nine of 12 companies calculated the percentage overage added into each study forecast; two in 12 used percentage added into aggregated study forecast; and one company calculated overage as a percentage of actual used (based on historical use).

	Additional data in Figure 6 revealed more insights into organizations’ attempts to manage waste or overage. As shown, 74% of packaged product was shipped to clinical sites and 90% of manufactured or procured product was packaged. However, with the aforementioned 67.8% of product shipped and dispensed, this reflects challenges and opportunities in the alignment of clinical supply and clinical study operation execution.

	The majority of respondents indicated that strategies did not differ between Europe and the U.S. for sourcing temperature maintenance products (e.g., shipping containers). Temperature monitoring strategies were primarily electronic. From the survey data gathered on temperature excursions occurring in transit were more frequent at customs clearance or at a site. The time taken to perform disposition of a shipment excursion was 57 hours, on average, with a range of six to 114 hours.

	A number of other factors impact distribution, creating additional challenges for clinical supply executives. One challenge is the delay caused by obtaining import licenses-and the top countries listed were Argentina, Russia, China, Colombia and India. 

	Company approaches varied regarding the use of expiry dates. They were split on this issue, with five companies indicating their approach varied and six saying it did not. In addition, 11 companies indicated that they allow different expiry or “use-by” dates at different levels of the distribution chain; four reported they did not. Another area where companies had mixed approaches was regarding consistency in putting expiry dates on the shipping package. Nine organizations indicated they put expiry dates on a shipping package while seven revealed they did not. The primary methods for managing expiry date updates were to perform extension labelling at the site or to return to a depot to conduct it. 

	Challenges in managing returns of drugs and supplies are also evident in a few key areas, including reconciliation and document destruction, complexity of managing returns and regulatory requirements. Managing returns varied across companies and could be destroyed at a site, returned to a local depot by region or country or returned to a central location. 

	Key performance indicators of distribution success were on-time shipments, percent of shipments with temperature excursions, percent minor deviations and on-time release or product released for use at a site (see Figure 7 below). Other indicators used were percent major deviations and on-time receipt of shipments.

	Companies were split on use of pooled supplies, with nine reporting that they used pooled supplies, while seven did not. The top challenge to using pooled supplies is the regulatory variation in acceptance across countries. No companies report using e-labelling, as regulations regarding its usage were expected to be clarified in the middle of this year.

	The results of this study reveal that a great number of supplies are shipped across the globe. Of 73 global clinical studies reaching 5,682 sites, there were, on average 1,538 mean shipments. Companies used a mixed approach of centralized, local and regional strategies for shipments. Shipping times to investigative sites took an average of 3.4 days, with a coefficient of variation of 1.4, indicating disparities among shipping times to global sites (Figure 6).

	Our findings for product impact and overage can help identify areas for continued improvement regarding cost and quality issues within clinical supply. Of the 57 studies further analyzed, there were 19 errors in shipping and handling and nine with site mishandling. This result represented 3% of studies and can be examined in further detail. Within the shipping and handling category, errors with temperature excursion were included. Any clinical supply logistic error rate may translate to potential site stock-out or product quality impact of the treatment that study subjects are waiting for. Companies can potentially look more closely into errors with product impact to see if there are strategies that can be implemented either within organizations or at the site level to reduce such instances. It is also critical for companies to achieve a balance between the risks and costs to optimize the clinical supply chain and increase quality. While cost containment is a priority, reducing inefficiencies in the supply chain is an ongoing challenge.

	The results of this study indicate that two-thirds of all product shipped to sites was actually dispensed to patients. This result is considered typical given anecdotal evidence, with some organizations potentially having higher rates of overage as suggested by published data on this topic.

 Coverage for geographic spread is another factor. It is acknowledged that waste varies by study, but the amount of overage impacts both cost and efficiency of supply. In addition, there is a

 relationship between clinical supply availability and the success of a study. A number of other factors such as cross-collaboration with clinical operations and other functions should be considered, especially those that manage the security of the supply. Companies may need to increase their “safety” stocks to address all the risks and potential scenarios that may occur. 

	Cross-collaboration also plays a role in the distribution strategy that an organization adopts. The studies we analyzed were split among centralized, local and regional strategies. Strategies varied across studies and by organization, perhaps due to the wide fragmentation and varied infrastructure of the global markets involved, but also due to unique company practices, set-up and strategies. It would require further study to examine what the drivers are for implementation of specific strategies and how best to optimize their usage. Nearly 60% of the trials we examined utilized outsourced logistics in managing the clinical supply chain, indicating that coordination among partners and vendors is also an important part of managing costs and efficiencies. 

	Forecasting supply can be linked to use of technology. Being able to forecast supply and the impact of factors such as enrollment, site selection and productivity and country selection is critical. Compared to 10 years ago, IRT plays a larger role in various clinical supply processes, from drug ordering and randomization to data extension management and drug reconciliation. Companies in our study estimated that nearly 70% of their studies are using IRT systems. The use of IRT is also evolving and taking on new functionality and standard platforms, as well as being integrated with other systems. Necessity and dependence of IRT becomes ever more critical to meet today’s clinical study complexity, improve study efficiency and enhance compliance. Many organizations have already implemented cross-functional IRT governance teams to align processes within the organization and with vendors. 

	The results of this study provide a useful set of baseline measures for clinical supply professionals. The major findings also suggest an even greater need for upfront planning, risk mitigation, reducing waste, increasing efficiency and promoting cross-collaboration among those managing or involved with drug supply. In addition, there may be missed opportunities for cost savings by reexamining approaches to overage. There are also potential opportunities to improve the quality of product impact during shipping and handling. Furthermore, the relationship of clinical study efficiency with clinical supply logistics shall be studied by identifying and correlating the key strategy designs and key performance indicators.

, PhD, is Senior Research Fellow, Tufts CSDD, Tufts University, email: 

, is Senior Director, Clinical Trial Materials Management, Sunovion Pharmaceuticals Inc.; 

, PharmD, is Director, Clinical Pharmacy, Astellas US Technologies Inc.; 

,  PhD, MBA, is Global Director, Strategic Development Solutions, Clinical Supply Services, Catalent Pharma Solutions; 

, MBA, is Director of Sponsored Research Programs, Tufts CSDD

To whom all correspondence should be addressed

	1. Getz KA, Brown C. High turnover, protocol noncompliance plague the global site landscape. Tufts Center for the Study of Drug Development Impact Report, 17(1) 1-3 (January/February 2015). 

	2. Eighth UPS Pain in the Supply Chain Healthcare Survey. 

https://www.ups.com/media/en/UPS-PITC-Executive-Summary-North-America.pdf

	3. Survey: Supply chain execs to up clinical trial logistics spend. 

http://www.outsourcing-pharma.com/Clinical-Development/Survey-Supply-chain-execs-to-up-clinical-trial-logistics-spend

	4. Getz KA, Kim J, Stergiopoulos S, Kaitin KI. New governance mechanisms to optimize protocol design. 

Therapeutic Innovation & Regulatory Science

	5. Getz KA, Stergiopoulos S, Marlborough M, Whitehill J, Curran M, Kaitin KI. Quantifying the magnitude and cost of collecting extraneous protocol data

	6. Lamberti MJ, Costello M, and Getz K. Global Supply Chain Management. From tactical to strategic: Tracking the evolution of global clinical supply chain management.

	7. Clinical Trials – Regulation EU 5 November 2013. 

http://ec.europa.eu/health/files/eudralex/vol-1/2013_c343_01/2013_c343_01_en.pdf

	8. Mcharg J. Ensuring the Integrity of Temperature-Sensitive Drugs along the Cold Chain, Pharma’s Almanac, Global Pharmaceutical Supply Chain Trends Q4 2015 edition, (A Nice Insight Supplement), 2-48 (2015). 

	9. EMA Delivery time frame for the EU portal and EU database. 

http://www.ema.europa.eu/docs/en_GB/document_library/Other/2015/12/WC500199078.pdf

	10. Brown D. Trends in Contract Pharma Serialization and Supply Chain Security.

	11. Lamberti MJ, Walsh T, Getz KA. Tracking trial cost drivers: the impact of comparator drugs and co-therapies. 

http://www.pharmexec.com/tracking-trial-cost-drivers-impact-comparator-drugs-and-co-therapies

Articles

Study collects first comprehensive metrics on current supply management and distribution practices.

Christine Milligan, PhD

Articles

Assessing Global Clinical Supply Logistics